- Split View
-
Views
-
Cite
Cite
Sally Robyn Isberg, Scott Maxwell Johnston, Yizhou Chen, Christopher Moran, First Evidence of Higher Female Recombination in a Species with Temperature-Dependent Sex Determination: the Saltwater Crocodile, Journal of Heredity, Volume 97, Issue 6, November/December 2006, Pages 599–602, https://doi.org/10.1093/jhered/esl035
- Share Icon Share
Abstract
The first evidence of genetic linkage and sex-specific recombination in the order Crocodylia is reported. This study was conducted using a resource pedigree of saltwater crocodiles consisting of 16 known-breeding pairs (32 adults) and 101 juveniles. A total of 21 microsatellite loci were available for analysis. Ten of the 21 loci showed linkage with 4 linkage groups: 3 pairwise (Cj131/Cj127, CUD68/Cj101, and Cj107/Cp10) and 1 four-locus (Cj122, CUD78, Cj16, and Cj104) being found. Linkage analysis on the 21 loci revealed evidence of sex-specific differences in recombination rates. All 5 nonzero interlocus intervals were longer in females than in males, with the 4-loci linkage group 3-fold longer in females than in males (41.63 cM and 14.1 cM, respectively). This is the first report of sex-specific recombination rates in a species that exhibits temperature-dependent sex determination.
The Australian crocodile industry produces saltwater crocodile (Crocodylus porosus) skins for the luxury fashion industry (Isberg, Thomson, et al. 2004). The industry is still in its infancy, and as such a genetic improvement program for the Australian saltwater crocodile industry has only recently been published (Isberg, Thomson, et al. 2004; Isberg et al. 2005a, 2005b; Isberg, Thomson, Nicholas, Barker, et al. 2006; Isberg, Thomson, Nicholas, Webb, et al. 2006). This involved estimating breeding values for candidate saltwater crocodiles using phenotypic records for economically important traits. Continuing on from this research, if quantitative trait loci (QTL) could be mapped, marker-assisted selection would reduce the generation interval on crocodile farms by expediting the identification of breeding replacements, thus enhancing production efficiency (Isberg, Chen, et al. 2004; Isberg, Thomson, et al. 2004). However, before a QTL map can be developed, a genetic linkage map must be established using microsatellite makers and physically mapped onto the chromosomes using techniques such as fluorescent in situ hybridization.
Genomic research in the order Crocodylia is currently limited to species-specific karyotypes, diploid (2n) chromosome numbers, and chromosome banding maps (Cohen and Gans 1970; King et al. 1986; Chavananikul et al. 1998). For example, in the saltwater crocodile 2n = 34 (Cohen and Gans 1970; Chavananikul et al. 1998), and the described karyotype includes 4 pairs of large metacentric, 5 pairs of acrocentric, 3 pairs of submetacentric, and 5 pairs of small metacentric chromosomes (Chavananikul et al. 1998). Chromosome banding maps have been developed for the saltwater crocodile to the point of establishing band patterns of nucleolar organizer regions (Chavananikul et al. 1998). No estimates of total genome size, descriptions of telomere dynamics, or synapsis initiation patterns have yet been described.
In this study, we report the first evidence of genetic linkage in a crocodilian species. In addition, the first evidence of sex-specific linkage in a species with temperature-dependent sex determination (TSD) is reported, in each case finding lower recombination in males.
Materials and Methods
The resource pedigree used in this study consisted of 133 individuals from 16 known family groups from Janamba Croc Farm (Northern Territory, Australia), as described in Isberg, Chen, et al. (2004). Thirty-two wild-caught adults were maintained as known-breeding pairs in unitized (one male and one female) pens. There was an average of 6.3 progeny per breeding pair, and their pedigree was confirmed in Isberg, Chen, et al. (2004). Blood sampling and DNA extraction techniques are described in Isberg, Chen, et al. (2004).
Experimental Protocol
Isberg, Chen, et al. (2004) presented results from 15 microsatellite markers. Of the 15 markers used in this study, 1 (C391) did not amplify whereas 3 (Cj35, CUJ-131, and CU4-121) were evaluated on the adults only. Therefore, the purpose of this study was to complete the evaluation of Cj35, CUJ-31, and CU4-121 on the progeny as well as evaluate 10 additional microsatellites on this resource pedigree (FitzSimmons et al. 2001). The primer sequences for the 13 microsatellites are shown in Table 1.
HW P value | Informative meiosis | |||||||
Locus | Primer sequences | N | k | HO | HE | Male | Female | |
Cj107 | F ACCCCGCATTCTGCCAAGGTG | 29 | 5 | 0.483 | 0.526 | 0.34 | 26 | 69 |
R GTTTATTGCCATCCCCACTGTGTC | ||||||||
Cj35 | F GTTTAGAAGTCTCCAAGCCTCTCAG | 30 | 2 | 0.333 | 0.505 | 0.13 | 61 | 39 |
R CTGGGGCAAGGATTTAACTCTC | ||||||||
CU4-121 | F GGTCAGCTAGCAGGGTG | 32 | 3 | 0.625 | 0.467 | 0.05 | 47 | 73 |
R TGGGGAAATGATTATTGTAA | ||||||||
CU5-123 | F GGGAAGATGACTGGAAT | 30 | 3 | 0.267 | 0.272 | 1.00 | 33 | 16 |
R AAGTGATTAACTAAGCGAGAC | ||||||||
CUD78 | F GAAGTGAATGCCATCTATCA | 30 | 3 | 0.500 | 0.646 | 0.10 | 51 | 49 |
R AATTGCATCCCCTTTTG | ||||||||
CUI-99.2 | F CACTGTGGGGGCCTCAATCTG | 31 | 2 | 0.225 | 0.204 | 1.00 | 14 | 28 |
R AGGCAGGTGGTAGGACCCTAGCAAT | ||||||||
CUJ-131 | F GTCCCTTCCAGCCCAAATG | 32 | 3 | 0.250 | 0.226 | 1.00 | 32 | 23 |
R CGTCTGGCCAGAAAACCTGT | ||||||||
CUJB-131 | F CCTGCCCAAGCCCATCAAT | 32 | 2 | 0.031 | 0.062 | 1.00 | 5 | 0 |
R CCCTTTTGGCATGGCACAGT |
HW P value | Informative meiosis | |||||||
Locus | Primer sequences | N | k | HO | HE | Male | Female | |
Cj107 | F ACCCCGCATTCTGCCAAGGTG | 29 | 5 | 0.483 | 0.526 | 0.34 | 26 | 69 |
R GTTTATTGCCATCCCCACTGTGTC | ||||||||
Cj35 | F GTTTAGAAGTCTCCAAGCCTCTCAG | 30 | 2 | 0.333 | 0.505 | 0.13 | 61 | 39 |
R CTGGGGCAAGGATTTAACTCTC | ||||||||
CU4-121 | F GGTCAGCTAGCAGGGTG | 32 | 3 | 0.625 | 0.467 | 0.05 | 47 | 73 |
R TGGGGAAATGATTATTGTAA | ||||||||
CU5-123 | F GGGAAGATGACTGGAAT | 30 | 3 | 0.267 | 0.272 | 1.00 | 33 | 16 |
R AAGTGATTAACTAAGCGAGAC | ||||||||
CUD78 | F GAAGTGAATGCCATCTATCA | 30 | 3 | 0.500 | 0.646 | 0.10 | 51 | 49 |
R AATTGCATCCCCTTTTG | ||||||||
CUI-99.2 | F CACTGTGGGGGCCTCAATCTG | 31 | 2 | 0.225 | 0.204 | 1.00 | 14 | 28 |
R AGGCAGGTGGTAGGACCCTAGCAAT | ||||||||
CUJ-131 | F GTCCCTTCCAGCCCAAATG | 32 | 3 | 0.250 | 0.226 | 1.00 | 32 | 23 |
R CGTCTGGCCAGAAAACCTGT | ||||||||
CUJB-131 | F CCTGCCCAAGCCCATCAAT | 32 | 2 | 0.031 | 0.062 | 1.00 | 5 | 0 |
R CCCTTTTGGCATGGCACAGT |
HW P value | Informative meiosis | |||||||
Locus | Primer sequences | N | k | HO | HE | Male | Female | |
Cj107 | F ACCCCGCATTCTGCCAAGGTG | 29 | 5 | 0.483 | 0.526 | 0.34 | 26 | 69 |
R GTTTATTGCCATCCCCACTGTGTC | ||||||||
Cj35 | F GTTTAGAAGTCTCCAAGCCTCTCAG | 30 | 2 | 0.333 | 0.505 | 0.13 | 61 | 39 |
R CTGGGGCAAGGATTTAACTCTC | ||||||||
CU4-121 | F GGTCAGCTAGCAGGGTG | 32 | 3 | 0.625 | 0.467 | 0.05 | 47 | 73 |
R TGGGGAAATGATTATTGTAA | ||||||||
CU5-123 | F GGGAAGATGACTGGAAT | 30 | 3 | 0.267 | 0.272 | 1.00 | 33 | 16 |
R AAGTGATTAACTAAGCGAGAC | ||||||||
CUD78 | F GAAGTGAATGCCATCTATCA | 30 | 3 | 0.500 | 0.646 | 0.10 | 51 | 49 |
R AATTGCATCCCCTTTTG | ||||||||
CUI-99.2 | F CACTGTGGGGGCCTCAATCTG | 31 | 2 | 0.225 | 0.204 | 1.00 | 14 | 28 |
R AGGCAGGTGGTAGGACCCTAGCAAT | ||||||||
CUJ-131 | F GTCCCTTCCAGCCCAAATG | 32 | 3 | 0.250 | 0.226 | 1.00 | 32 | 23 |
R CGTCTGGCCAGAAAACCTGT | ||||||||
CUJB-131 | F CCTGCCCAAGCCCATCAAT | 32 | 2 | 0.031 | 0.062 | 1.00 | 5 | 0 |
R CCCTTTTGGCATGGCACAGT |
HW P value | Informative meiosis | |||||||
Locus | Primer sequences | N | k | HO | HE | Male | Female | |
Cj107 | F ACCCCGCATTCTGCCAAGGTG | 29 | 5 | 0.483 | 0.526 | 0.34 | 26 | 69 |
R GTTTATTGCCATCCCCACTGTGTC | ||||||||
Cj35 | F GTTTAGAAGTCTCCAAGCCTCTCAG | 30 | 2 | 0.333 | 0.505 | 0.13 | 61 | 39 |
R CTGGGGCAAGGATTTAACTCTC | ||||||||
CU4-121 | F GGTCAGCTAGCAGGGTG | 32 | 3 | 0.625 | 0.467 | 0.05 | 47 | 73 |
R TGGGGAAATGATTATTGTAA | ||||||||
CU5-123 | F GGGAAGATGACTGGAAT | 30 | 3 | 0.267 | 0.272 | 1.00 | 33 | 16 |
R AAGTGATTAACTAAGCGAGAC | ||||||||
CUD78 | F GAAGTGAATGCCATCTATCA | 30 | 3 | 0.500 | 0.646 | 0.10 | 51 | 49 |
R AATTGCATCCCCTTTTG | ||||||||
CUI-99.2 | F CACTGTGGGGGCCTCAATCTG | 31 | 2 | 0.225 | 0.204 | 1.00 | 14 | 28 |
R AGGCAGGTGGTAGGACCCTAGCAAT | ||||||||
CUJ-131 | F GTCCCTTCCAGCCCAAATG | 32 | 3 | 0.250 | 0.226 | 1.00 | 32 | 23 |
R CGTCTGGCCAGAAAACCTGT | ||||||||
CUJB-131 | F CCTGCCCAAGCCCATCAAT | 32 | 2 | 0.031 | 0.062 | 1.00 | 5 | 0 |
R CCCTTTTGGCATGGCACAGT |
For every microsatellite locus, the amplification reaction took place in a total volume of 15 μl. Polymerase chain reaction (PCR) reagents included 1 unit of Taq DNA polymerase (various sources), 1× PCR buffer (Promega, Madison, WI), and final concentrations of 0.1 mM deoxynucleotide triphosphates, 20 pmol each of forward and reverse primer, 0.6–2.0 mM MgCl2, and approximately 50–200 ng of template DNA. Standard PCR conditions included a touchdown protocol with an initial denaturation at 95 °C for 15 min; followed by 3 cycles of 95 °C for 40 sec, 63 °C for 1 min, and 72 °C for 1 min 30 sec; followed by 5 cycles of 95 °C for 40 sec, 61 °C for 1 min, and 72 °C for 1 min 30 sec; followed by 35 cycles of 95 °C for 40 sec, 59 °C for 1 min, and 72 °C for 1 min 30 sec; and finally being held at 72 °C for 20 min. PCR products and size standard (GeneScan™ 500 Tamra™, Applied Biosystems, Inc, Foster City, CA) were either run on a denaturing 6% polyacrylamide gel using an ABI 373 sequencer (Applied Biosystems, Inc) or a capillary-based ABI PRISM® 3700 DNA Analyzer (Applied Biosystems, Inc). Alleles were visually scored using Genotyper software (Applied Biosystems, Inc).
Microsatellite and Linkage Analysis
The number of alleles and null alleles for each locus as well as observed and expected heterozygosities were calculated using CERVUS 2.0 (Marshall et al. 1998). Tests for Hardy–Weinberg equilibrium and linkage disequilibrium were conducted using ARLEQUIN 2.000 (Schneider et al. 2000). Adjustment for multiple testing was carried out using Bonferroni's correction (Bonferroni 1936).
Linkage analyses were performed using CRIMAP version 5.0 (Green et al. 1990). Pairwise linkage analysis identified linkages between any two loci, assuming both equal and unequal recombination rates in the 2 sexes. Pairwise linkage was considered significant if the logarithm of odds (LOD) score was greater than 3. Linkage groups with three or more linked loci were ordered using the ALL option, whereas the FLIPS option showed the framework order of the loci. Finally, the BUILD option (multipoint analysis) was used to estimate the recombination rates and Kosambi mapping distances for the female, male, and sex-averaged map. To test the significance of the sex-averaged versus sex-specific linkage groups, the difference in the LOD scores was multiplied by 4.6 to obtain a chi-square value and compared with a chi-square distribution using degrees of freedom (df) equal to the number of free intervals between markers (Ott 1991). Linkage maps were drawn using MAPCHART 2.1 (Voorrips 2002).
Results and Discussion
Of the 13 microsatellites evaluated, 10 were genotyped on all available samples whereas 3 failed to amplify (CR52, CUI-108, and Cj128). From these 10 microsatellites, 8 were informative (Table 1) whereas 2 were monomorphic (CUC-20 and pCp-8H4) and omitted from further analyses. Combined with the 11 microsatellites from Isberg, Chen, et al. (2004), all genotype frequencies conformed to expectations (P > .05) with the exception of Cj104 (P = .028; Isberg, Chen, et al. 2004), Cj101 (P = .048; Isberg, Chen, et al. 2004), and CU4-121 (P = .045; this study). Application of Bonferroni's correction for multiple testing (Rice 1989) indicated that these differences were no longer significant (P > .05). Thus, these loci were unlikely to bias likelihood estimates and were therefore kept in the analysis. Tests of null alleles were conducted in CERVUS, and there was no evidence of null allele segregation.
Using the 21 microsatellite markers, we found 4 linkage groups. Three were pairwise linkages (Cj127 and Cj131, Cj101 and CUD68, Cj107 and Cp10), whereas the fourth linkage group consisted of 4 markers (Cj16, Cj104, Cj122, and CUD78; Table 2). Figure 1 shows the map order for the 4-marker linkage group. The female genetic map is 3-fold longer than the male map, with the loci Cj122 and Cj16 appearing to be immediately adjacent to each other (sex-averaged and sex-specific recombination = 0; LOD 7.53; Table 2). The statistical support for the map order between loci Cj122/Cj16 and CUD78 is just below significance (LOD 2.56), whereas the interval between CUD78 and Cj104 is significant (LOD 4.06; Figure 1).
Linkage group | Linkage interval | Sex averaged | Sex specific | |||||
r | LOD | Female r | Male r | LOD | χ12 | χ22 | ||
LG1 | Cj127–Cj131 | 0.19 | 6.97 | 0.25 | 0.09 | 8.24 | 5.85* | |
LG2 | Cj101–CUD68 | 0.06 | 8.78 | 0.16 | 0.00 | 10.58 | 8.29** | |
LG3 | Cj107–Cp10 | 0.14 | 3.29 | 0.22 | 0.00 | 4.76 | 6.77** | |
LG4 | Cj122–Cj16 | 0.00 | 7.53 | 0.00 | 0.00 | 7.53 | — | 6.26* |
Cj16–CUD78 | 0.10 | 8.09 | 0.15 | 0.06 | 8.76 | 3.09ns | ||
Ci122–CUD78 | 0.00 | 3.31 | 0.00 | 0.00 | 3.31 | — | ||
CUD78–Cj104 | 0.12 | 6.52 | 0.25 | 0.03 | 7.75 | 5.66* |
Linkage group | Linkage interval | Sex averaged | Sex specific | |||||
r | LOD | Female r | Male r | LOD | χ12 | χ22 | ||
LG1 | Cj127–Cj131 | 0.19 | 6.97 | 0.25 | 0.09 | 8.24 | 5.85* | |
LG2 | Cj101–CUD68 | 0.06 | 8.78 | 0.16 | 0.00 | 10.58 | 8.29** | |
LG3 | Cj107–Cp10 | 0.14 | 3.29 | 0.22 | 0.00 | 4.76 | 6.77** | |
LG4 | Cj122–Cj16 | 0.00 | 7.53 | 0.00 | 0.00 | 7.53 | — | 6.26* |
Cj16–CUD78 | 0.10 | 8.09 | 0.15 | 0.06 | 8.76 | 3.09ns | ||
Ci122–CUD78 | 0.00 | 3.31 | 0.00 | 0.00 | 3.31 | — | ||
CUD78–Cj104 | 0.12 | 6.52 | 0.25 | 0.03 | 7.75 | 5.66* |
Linkage group | Linkage interval | Sex averaged | Sex specific | |||||
r | LOD | Female r | Male r | LOD | χ12 | χ22 | ||
LG1 | Cj127–Cj131 | 0.19 | 6.97 | 0.25 | 0.09 | 8.24 | 5.85* | |
LG2 | Cj101–CUD68 | 0.06 | 8.78 | 0.16 | 0.00 | 10.58 | 8.29** | |
LG3 | Cj107–Cp10 | 0.14 | 3.29 | 0.22 | 0.00 | 4.76 | 6.77** | |
LG4 | Cj122–Cj16 | 0.00 | 7.53 | 0.00 | 0.00 | 7.53 | — | 6.26* |
Cj16–CUD78 | 0.10 | 8.09 | 0.15 | 0.06 | 8.76 | 3.09ns | ||
Ci122–CUD78 | 0.00 | 3.31 | 0.00 | 0.00 | 3.31 | — | ||
CUD78–Cj104 | 0.12 | 6.52 | 0.25 | 0.03 | 7.75 | 5.66* |
Linkage group | Linkage interval | Sex averaged | Sex specific | |||||
r | LOD | Female r | Male r | LOD | χ12 | χ22 | ||
LG1 | Cj127–Cj131 | 0.19 | 6.97 | 0.25 | 0.09 | 8.24 | 5.85* | |
LG2 | Cj101–CUD68 | 0.06 | 8.78 | 0.16 | 0.00 | 10.58 | 8.29** | |
LG3 | Cj107–Cp10 | 0.14 | 3.29 | 0.22 | 0.00 | 4.76 | 6.77** | |
LG4 | Cj122–Cj16 | 0.00 | 7.53 | 0.00 | 0.00 | 7.53 | — | 6.26* |
Cj16–CUD78 | 0.10 | 8.09 | 0.15 | 0.06 | 8.76 | 3.09ns | ||
Ci122–CUD78 | 0.00 | 3.31 | 0.00 | 0.00 | 3.31 | — | ||
CUD78–Cj104 | 0.12 | 6.52 | 0.25 | 0.03 | 7.75 | 5.66* |
These are the first reported genetic linkage groups in any reptile. In all cases involving nonzero recombination, the recombination rate is higher in females, and for all 4 linkage groups, a sex-specific model of recombination fits the data significantly better than a model assuming equal recombination in males and females.
These are also the first genetic linkages reported for a species with TSD. Crocodilians, like some species of turtles and lizards, do not have sex chromosomes (Cohen and Gans 1970; Sarre et al. 2004), but rather their sex is determined by the incubation temperature (Lang and Andrews 1994). Although sex-determining incubation temperatures and thermosensitive periods vary between crocodilian species, an incubation temperature of 32 °C for C. porosus produces 86% males, whereas a 1 °C deviation decreases markedly the proportion of males produced (16% at 31 °C and 17% at 33 °C; Lang and Andrews 1994).
The interesting feature of this finding is the further exception to the Haldane–Huxley rule (Haldane 1922; Ott 1991), which predicted that recombination would be less frequent in the heterogametic sex. There are already sufficient exceptions at the level of species (for example, the tammar wallaby [Zenger et al. 2002] and the great reed warbler [Hansson et al. 2005]) and individual chromosomes to challenge the general validity of the Haldane–Huxley rule (Moran and James 2005). The higher frequency of female recombination in C. porosus is a further exception to this rule because of the lack of sex chromosomes suggesting that it might be some aspect of the timing, duration, or other biological features of female meiosis that is responsible for the general tendency for elevated female recombination and shows clearly that sex chromosomes and genetic sex determination have nothing to do with it.
This research was approved by the University of Sydney Animal Ethics Committee (reference number N00/10-2001/3/3442). This research was supported by Rural Industries Research and Development Corporation in collaboration with Janamba Croc Farm, Northern Territory, Australia. S.R.I. was supported by a University of Sydney Postgraduate Award (cofunded). We thank Mr Stuart Barker at Janamba Croc Farm for allowing access to samples for this study and Professor Chris Haley for statistical advice.
References
Author notes
Corresponding Editor: William Modi